The goal of this R03 proposed investigation is to identify prospectively the individual and contextual factors that predict and can ultimately prevent extreme aggression exhibited by youth during institutional treatment. The proposed study uses existing data archives to address this broad goal by 1) Accessing and coding individual, contextual, and behavioral data from archived sources (intake/treatment records and a critical incident database) in an inpatient youth psychiatric facility; and 2) Testing a model of aggressive behavior occurring during institutionalized care by examining the prediction of self- and other-directed aggressive behaviors and related critical incidents (i.e., seclusions and restraints) from individual and contextual variables. This investigation is guided by several specific aims. First, to identify the individual biopsychological characteristics and contextual socializing influences that are associated with the occurrence of self-directed and other-directed aggression during institutional treatment. This involves identifying factors in a very high-risk sample of youth that discriminate those who engage in serious aggression during institutional treatment from those who do not. Second, to examine the extent to which individual and contextual factors predict the patterns and persistence of aggression throughout institutional treatment. This involves the study of how individual-contextual factors lead to increases, decreases, or non-linear trends in aggression during treatment. Information about individual-contextual predictors of self- and other-directed aggressive behaviors derived through the two sets of analyses will be used to develop assessment measures suitable for research and practice on the development and treatment of aggression in very high-risk youth. This instrument will be designed to serve as the basis for a larger scale longitudinal study of aggression in that population, and also will be directly applicable to treatment planning for individual youth. A critical focus of this investigation also is to specify a model of aggression in very high-risk youth that can be used to facilitate the dissemination of findings and spur additional research. Results from the proposed study will have implications for the host facility as well as to the larger audience of scientists and practitioners invested in the reduction and prevention of assaultive and self-injurious behaviors among high-risk children and adolescents.